Acute Physiology, Age, Chronic Health Evaluation Research Articles

Relationships among Demographic, Clinical, and Psychological Factors Associated with Family Caregiver Readiness to Participate in Intensive Care Unit Care.

Rationale: There has been a paradigm shift to partner with family caregivers by actively involving them in the direct care of the patient throughout the critical illness trajectory. Before effectively engaging family members in patient care, clinicians must assess characteristics and circumstances that may affect caregiver readiness to assume a caregiving role in the intensive care unit (ICU). Objectives: To determine how demographic, clinical, and psychological factors are related to characteristics of family caregiver readiness to engage in ICU patient care. Methods: A convenience sample of ICU family caregivers of both adult and pediatric patients in the ICU was recruited for this cross-sectional study. Participants completed the following measures: PROMIS-29 (Patient-Reported Outcomes Measurement and Information System); HADS (Hospital Anxiety and Depression Scale); CaSES (Caregiver Self-Efficacy Scale); Prep Scale (Preparedness for Caregiving Scale); Patient Activation Measure for Caregivers; and FCMFHS (Family Caregiver's Motives for Helping Scale). Data were collected via self-report at a single time point while the caregiver was visiting the critically ill patient in the ICU. Data analysis consisted of descriptive statistics and bivariate correlations. Results: Caregivers (N = 127) were primarily White (82.7%), females (77.2%), with a mean age of 51.8 (standard deviation [SD], 15.6). Most were either spouses (37.8%) or parents (32.3%) of the patient in the ICU. Patients were primarily adult (76.4%) with a mean APACHE (Acute Physiology, Age, Chronic Health Evaluation) III of 45.9 (SD, 22.5). There were significant (P < 0.05) negative correlations between depression, anxiety, and fatigue and all subscales of self-efficacy (resilience r = -0.18 to -0.30; self-maintenance r = -0.44 to -0.63; emotional connectivity r = -0.27 to -0.41; instrumental giving r = -0.34 to -0.46). Caregiver depression was negatively correlated with caregiver activation (r = -0.199) and caregiver preparedness (r = -0.300). Social satisfaction was positively correlated (P < 0.05) with caregiving preparedness, motivation, and all subscales of self-efficacy (preparedness r = 0.19; motivation r = 0.24; resilience r = 0.21; self-maintenance r = 0.49; emotional connectivity r = 0.29; instrumental giving r = 0.36). Conclusions: We found that caregiver symptoms of depression, anxiety, and fatigue are inversely related to caregiver preparation, motivation, and self-efficacy. To develop effective interventions for ICU family caregivers, further research is needed to understand the relationship between caregiver well-being, caregiving readiness, and caregiver involvement in patient care.

The Effect of Organ System Surgery on Intensive Care Unit Mortality in a Cohort of Critically Ill Surgical Patients.

The critical illness burden in the United States is growing with an aging population obtaining surgical intervention despite age-related comorbidities. The effect of organ system surgical intervention on intensive care units (ICUs) mortality is unknown. We performed an 8-year retrospective analysis of surgical ICU patients. Poisson regression analysis was performed assessing the relative risk of in-hospital mortality based on surgical intervention. Of 468000 ICU patients included, 97968 (20.9%) were surgical admissions and 97859 (99.9%) had complete outcomes data. Nonsurvivors were older (68.8 ± 15.4 vs. 62.7 ± 15.8years, P < .001) with higher Acute Physiology, Age, Chronic Health Evaluation (APACHE) III Scores (81.4 ± 33.6 vs. 46.7 ± 20.1, P < .001. Patients with gastrointestinal (GI) (n = 1,558, 7.8%), musculoskeletal (n = 277, 5.5%), and neurological (n = 884, 4.6%) system operations had the highest mortality. Upon Poisson regression model, patients undergoing emergent operative interventions on the neurologic system (RR 1.86, 95% CI 1.67-2.07, P < .001) had increased relative risk of mortality when compared to emergent operative interventions on the cardiovascular system after controlling for pertinent covariates. Elective operative interventions on the respiratory (RR 2.39, 95% CI 2.03-2.80, P < .001), GI (RR 2.34, 95% CI 2.10-2.61, P < .001), and skin and soft tissue (RR 2.26, 95% CI 1.77-2.89, P < .001) systems had increased risk of mortality when compared to elective cardiovascular system surgery after controlling for pertinent covariates. We found significant differences in the risk of mortality based on organ system of operative intervention. The prognostication of critically ill patients undergoing surgical intervention is currently not accounted for in prognostic scoring systems.

Ventilation bundle compliance in two Australian intensive care units: An observational study

BackgroundThe ventilation bundle has been used in adult intensive care units to decrease harm and improve quality of care for mechanically ventilated patients. The ventilation bundle focuses on prevention of specific complications of mechanical ventilation; ventilator-associated pneumonia, sepsis, barotrauma, pulmonary oedema, pulmonary embolism, and acute respiratory distress syndrome. The Institute for Healthcare Improvement ventilation bundle consists of five structured evidence-based interventions: head of the bed elevation at 30–45°; daily sedation interruptions and assessment of readiness to extubate; peptic ulcer prophylaxis; deep vein thrombosis prophylaxis; and daily oral care with chlorhexidine. ObjectivesThe objective of the study was to evaluate the use of the ventilation bundle in two intensive care units in Victoria, Australia. MethodsThis is a 3-month prospective observational study in two intensive care units. Patient medical records were reviewed on days 3, 4, and 5 of mechanical ventilation using a prevalidated ventilation bundle checklist. ResultsA total of 96 critically ill patients required mechanical ventilation for more than 2 d. Patients had a mean age of 64.50 y (standard deviation = 14.89), with an Acute Physiology, Age, Chronic Health Evaluation (APACHE) III mean score of 79.27 (standard deviation = 27.11). The mean ventilation bundle compliance rate was 88.3% on the three consecutive mechanical ventilation days (day 3 = 79.4%, day 4 = 91.1%, and day 5 = 96.7%). There was a statistically significant difference in the mean APACHE III score between patients who had head of bed elevation and those without head of bed elevation, on days 3 (p = <0.001) and 4 (p = 0.007). ConclusionThe ventilation bundle elements were used in Australian intensive care units. The likelihood of having all ventilation bundle elements on day 3 was low if the patient's APACHE III score was high. However, the ventilation bundle compliance rate increased with mechanical ventilation days.

Clinical features and outcomes of hypertriglyceridemia-induced acute pancreatitis: Propensity score matching analysis from a prospective acute pancreatitis registry

BackgroundHypertriglyceridemia (HTG) is a well-known cause of acute pancreatitis (AP) and elevation of serum triglycerides (TG) to ≥1000 mg/dl is strongly indicative of HTG-induced AP (HTG-AP). HTG-AP is potentially associated with persistent organ failure and poor prognosis. Here, we compared differences in clinical features and outcomes between patients with HTG-AP and patients with AP due to other causes. MethodsA prospective AP registry was constructed in Gil Medical Center between June 2014 and May 2018. In total, 499 patients with AP were included for whom serum TG data at admission were available. ResultsHTG-AP was present in 52 patients (10.4%); these patients were younger than patients with AP due to other causes (39.62 ± 10.12 vs. 51.62 ± 17.41, p < 0.001). After propensity score matching adjusted by age, the factors associated with severity were more common in the HTG-AP group; these factors included the presence of systemic inflammatory response syndrome, Ranson’s score ≥3, acute physiology, age, chronic health evaluation (APACHE) II score ≥8 at admission, and C-reactive protein level >10 mg/dl after 24 h of hospitalization. There were no significant differences in complications or severity based on the revised Atlanta classification 2012. In addition, recurrence was more frequent in the HTG-AP group (25.0% vs. 6.4%, p < 0.001). ConclusionHTG-AP occurred in younger patients and showed more frequent recurrences than AP with other causes. Although factors related to severe feature were more common in HTG-AP during early phase, overall severity and prognosis were not different between the two groups.

The duration of mechanical ventilation in patients with chronic obstructive pulmonary disease and acute respiratory distress syndrome admitted to the intensive care unit: Epidemiological findings from a tertiary hospital

Context: Data for examining the duration and risk factors associated with the length of hospital stay with mechanical ventilation (MV) in patients with acute respiratory distress syndrome (ARDS) and chronic obstructive pulmonary disease (COPD) are scarce. Aims: The aim of the study was to examine the hospital stay duration with MV in such patients in the intensive care unit (ICU). Settings and Design: This cross-sectional study was conducted at King Abdulaziz Medical City in Riyadh between 2016 and 2018. Subjects and Methods: Adult patients with ARDS or COPD admitted to the ICU and receiving MV were included in the study. Their medical records were reviewed for information regarding the length of hospital stay, demographic profile, and comorbidity. Results: In total, 136 and 95 patients with ARDS and COPD, respectively, were admitted and received MV. The mean (standard deviation) age of patients with ARDS and COPD was 53 (19) and 68 (12) years, respectively. The patients with COPD had a higher obesity rate (61% in the obese category), higher mean Acute Physiology, Age, Chronic Health Evaluation (APACHE) II score (23.5 vs. 22.7), and preexisting chronic respiratory disease than those with ARDS. The median (interquartile range) MV duration in patients with ARDS was 7 (2–13) days, which was higher than that in patients with COPD (4 [1–9] days). Conclusions: Age, APACHE II score, and preexisting comorbidity contributed to the increase in the length of hospital stay with MV among patients with ARDS. Clinicians must consider such factors before using MV to improve patient outcomes.

Using bioimpedance analysis to assess intensive care unit patients with sepsis in the post-resuscitation period: a prospective multicentre observational study.

Clinicians lack well-validated, non-invasive, objective tools to guide volume management in the post-resuscitative period. Bioimpedance analysis (BIA) represents a novel method for guiding fluid management. We studied the relationship of BIA vector length (VL), an indicator of volume status, to the need for mechanical ventilation in patients with sepsis. This is a multicentre prospective observational study at four Canadian ICUs. We examined adult patients admitted to the ICU within 72 hr of a sepsis diagnosis. Patients underwent daily BIA measurements for 30 days, until discharge from the ICU, or until death. Our primary outcome was the ongoing need for invasive mechanical ventilation, and we examined the association with VL using a generalized estimating equation. Our secondary analyses were targeted to determine an association between VL and other measures of volume status and acute kidney injury (AKI). We enrolled 159 patients from four centres over 27 months. The mean (standard deviation [SD]) age was 64 (15) yr with a mean (SD) APACHE (acute physiology, age, chronic health evaluation) II score of 25 (10); 57% (n = 91) were male. A 50-unit (ohm·m) increase in VL over any time period was associated with a 30% decrease in the probability of requiring invasive mechanical ventilation (P < 0.03). Volume expansion, indicated by a shorter VL, correlated with higher edema scores (r = - 0.31; P < 0.001) and higher net 24-hr fluid balance (r = - 0.27, P < 0.001). Patients with AKI had a shorter overall VL (r = - 0.23; P = 0.003). An increase in VL over time is associated with a decrease in probability of requiring invasive mechanical ventilation. Vector length correlates with other commonly used volume assessment methods in post-resuscitation patients with sepsis.

Audit of Antibiotic Practices: An Experience from a Tertiary Referral Center.

ABSTRACTAimsTo estimate the prevalence of antibiotic de-escalation at admission in patients referred to a tertiary hospital in India. The secondary outcomes were the adequacy of empirical antibiotic therapy and culture positivity rates in the de-escalated group.Materials and methodsA prospective observational study, in a 20-bedded intensive care unit (ICU) of tertiary care hospital. Patients >18 years, surviving > 48 hours, were included (June– December 2017). Demographic data, previous cultures, and antibiotics from other hospitals, laboratory parameters in the first 24 hours, and severity of illness were noted. Changes made in antibiotic therapy within 48 hours were recorded. Patients were analyzed into three groups: “No change”–empiric therapy was maintained, “Escalation”–switch to or addition of an antibiotic with a broader spectrum, and “De-escalation”–switch to or interruption of a drug class.ResultsThe total number of patients eligible was 75. The mean age of the population is 43.38 (SD + 3.4) and groups were comparable in terms of mean sequential organ failure assessment score (SOFA) and acute physiology, age, chronic health evaluation (APACHE) 2. The prevalence of de-escalation was 60% at admission. The escalation group consisted of 24%. Sixteen percent patients belonged to no change group. Results showed that 38% of patients were on carbapenems, dual gram negative was given to 26%, and empirical methicillin-resistant staphylococcus aureus (MRSA) coverage was 28% on admission.ConclusionOur study aims to provide data about actual practices in the Indian scenario. It highlights the generous use of high-end antibiotics in the community. Indian practices are far cry from theoretical teaching and western data. The need for antibiotic stewardship program in our country for both public and private health sectors is the need of the hour.How to cite this articleSingh R, Azim A, Gurjar M, Poddar B, Baronia AK. Audit of Antibiotic Practices: An Experience from a Tertiary Referral Center. Indian Journal of Critical Care Medicine, January 2019;23(1):7-10.

Impact of individualized target mean arterial pressure for septic shock resuscitation on the incidence of acute kidney injury: a retrospective cohort study

BackgroundTo examine the relationship between delta mean arterial pressure (ΔMAP; MAP change between pre-admission minus post-resuscitation) and acute kidney injury (AKI) among patients with septic shock. In this retrospective, single-center cohort study of adult patients pre-admission MAP is defined as the median MAP recorded from 365 to 7 days before admission. Post-resuscitation MAP was median MAP during the 7th hour after initiating resuscitation.ResultsIn our cohort (N = 233; 55% male), the median (interquartile range [IQR]) age was 71 (58–81) years and the median (IQR) acute physiology, age, chronic health evaluation (APACHE) III score was 81 (66–97). Although those in the lowest ΔMAP quartile (−24.5 to 3.9 mmHg) had no demographic differences compared with the rest of the cohort, the odds ratio for AKI was 0.26 (95% CI 0.11–0.57) after adjustment for other known AKI risk factors. Among patients with a history of hypertension, the lowest quartile had an odds ratio for AKI of 0.12 (95% CI 0.04–0.37) after adjusting for risk factors for AKI in this cohort.ConclusionsThe incidence of AKI was lowest among those whose post-resuscitation MAP was closest to or higher than their pre-admission MAP. Further study regarding the effect of targeting the pre-admission MAP for post-resuscitation on the incidence of AKI is warranted.

RETRACTED ARTICLE: Evaluation of the Relationship Between Serum Vitamin D Level and Severity of Sepsis

Background: On the basis of the literature, vitamin D is known as an important medium in bodily immune function, and it therefore may play a role in the pathogenesis of sepsis. Objectives: In this study, we aimed to evaluate the relationship between vitamin D serum levels and sepsis severity. Patients and Methods: This study was a case-control study that evaluated adult patients admitted to the emergency department of Imam Reza hospital with suspected sepsis. These patients were enrolled in the study as the case group. In addition, healthy individuals without the sepsis diagnostic criteria were included in the control group. For all of the study participants, vitamin D levels were evaluated. The acute physiology age chronic health evaluation (APACHE) was used to evaluate disease severity in the case group. A difference of P < 0.05 was regarded as statistically significant. Results: A total of 112 patients were assessed: 56 in the control group and 56 in the case group. In the case group, 18 patients had sepsis, 25 patients had severe sepsis, and 13 patients were in septic shock. The mean ages of the patients in the case and control groups were 57.7 ± 15.15 and 58.6 ± 15.05 years, respectively (P = 0.741). Vitamin D levels in the case group were lower than in the control group (16.3 ± 10.7 versus 27.9 ± 11.46 ng/mL), and the difference between the groups was significant (P < 0.001). Mean vitamin D levels in the severe sepsis and septic shock groups were lower than in the sepsis group, and the mean level in the septic shock group was lower than in the severe sepsis group (P = 0.001). In the case group, there was a significant reverse correlation between APACHE II criteria and vitamin D levels (P < 0.001, r = -0.586). Conclusions: The results of this study indicated that patients with sepsis had lower serum vitamin D levels than healthy controls. Also, patients with more severe disease had lower serum vitamin D levels, but to evaluate causation and determine whether vitamin D supplementation could be effective in reducing the risk or severity of sepsis, randomized controlled trials should be conducted.

Trends in ICU Quality After Implementation of an Electronic Health Record

Trends in ICU Quality After Implementation of an Electronic Health Record

Vitamin D Insufficiency and Sepsis Severity in Emergency Department Patients With Suspected Infection

Vitamin D is increasingly recognized as an important mediator of immune function and may have a preventive role in the pathogenesis of sepsis. We sought to evaluate the association between vitamin D status and sepsis severity and hypothesized that vitamin D insufficiency would be associated with increased sepsis severity. This was a pilot study of emergency department (ED) patients age ≥ 18 years evaluated for suspected infection at an urban, teaching hospital. The authors measured illness severity using the following assessments at baseline and 24 hours: 1) severe sepsis, defined as suspected infection plus two or more elements of systemic inflammatory response syndrome criteria and acute dysfunction of one or more organ systems; 2) Acute Physiology Age Chronic Health Evaluation (APACHE) II scores; and 3) Sepsis-related Organ Failure Assessment (SOFA) scores. Vitamin D insufficiency was defined as baseline serum 25-hydroxyvitamin D (25OHD) levels <75 nmol/L. Eighty-one patients were enrolled, with a median age of 62 years (interquartile range [IQR] = 48-76 years), 47% were female, and 77% were white. At baseline, 64 (79%) had 25OHD levels of <75 nmol/L, and 43 (53%) had severe sepsis. At 24 hours, 48 (59%) had severe sepsis. Patients with baseline 25OHD levels of <75 nmol/L, compared to patients with 25OHD levels of ≥ 75 nmol/L, were more likely to have severe sepsis (61% vs. 24%; p = 0.006) and SOFA scores ≥ 2 (44% vs. 18%; p = 0.049). Additionally, at 24 hours, those with 25OHD levels of <75 nmol/L were more likely to have severe sepsis (67% vs. 29%; p = 0.005), dysfunction of two or more organ systems (50% vs. 18%; p = 0.02), APACHE II score of ≥ 25 (19% vs. 0%; p = 0.06), and SOFA scores of ≥ 2 (63% vs. 29%; p = 0.02). Additionally, all four patients who died during the index hospitalization had 25OHD levels of <75 nmol/L. Vitamin D insufficiency was associated with higher sepsis severity in ED patients hospitalized for suspected infection. Larger observational studies, mechanistic studies, and ultimately randomized controlled trials are needed to determine causation and to evaluate if vitamin D supplementation can reduce the risk of sepsis as a preventive or therapeutic strategy.

Current practice of management of bacteremic sepsis: a study in a tertiary care teaching hospital in Japan.

To investigate how patients with bacteremic sepsis are managed in a tertiary care teaching hospital. Prospective observational study on patients with bacteremic sepsis. Clinical and microbiological characteristics of bacteremic sepsis were analyzed in relation to prognosis. Severity of the illness was quantitatively analyzed by the APACHE (Acute Physiology, Age, Chronic Health Evaluation) III scoring system. Also investigated was how closely physicians paid attention to acute physiological alterations in patients. The 28-day mortalities in fifty hemodynamically stable patients and in twenty-three septic shock patients were 26% and 52%, respectively (p=0.028). Gram-positive organisms accounted for 54% of all organisms, with the mortality and incidence of septic shock being the same as with Gram-negative infections. The mean APACHE III score was 42.9 in survivors, and 76.5 in non-survivors (p < 0.001). Although serum levels of C-reactive protein and acute physiology score (APS) was significantly higher in non-survivors than in survivors, the correlation with APACHE III score was more prominent in APS. The number of vital signs recorded was 1.67 in physicians and 3.6 in nurses (p < 0.001). The present study proved that the APACHE III score accurately discriminates between survivors and non-survivors of patients with sepsis. By addressing the need for an objective evaluation of severity of illness, it strongly recommends that physicians should be made aware of physiologically defined sepsis and that they should pay closer attention to patients' physiological alterations to identify the development of sepsis in critically ill patients.

Adult respiratory distress syndrome in systemic lupus erythematosus: causes and prognostic factors: a single center, retrospective study.

To determine the causes and prognostic factors of Adult Respiratory Distress Syndrome (ARDS) in patients with systemic lupus erythematosus (SLE). Among 544 Korean SLE patients, who were followed in the Lupus Clinic of the Catholic Medical Center from 1993 to 1997, patients diagnosed as ARDS were examined retrospectively. During the study period, non-SLE patients with ARDS were investigated and then compared to SLE patients with ARDS in terms of clinical variables. Nineteen patients with SLE were revealed to have a history of ARDS (3.5%) and 13 (68.4%) of 19 patients died. Death related to ARDS was found in 34.2% of all deaths (n=38) from SLE during the study period. The frequency and causes of ARDS were as follows; 9 sepsis or bacteremia (47.4%), 2 miliary tuberculosis (10.5%), 2 invasive pulmonary aspergillosis (10. 5%), 2 acute pulmonary alveolar hemorrhage syndrome (10.5%), 1 acute lupus pneumonitis (5.3%), 1 massive hemorrhage due to placenta previa (5.3%), 1 aspiration pneumonitis (5.3%), 1 disseminated intravascular coagulation associated with systemic vasculitis (5.3%). The main organisms in sepsis were gram negative bacilli (61.5%) The median steroid dose administered 1 month before ARDS was significantly higher in patients (n=13) with infectious ARDS compared to those (n=6) with ARDS due to other causes (P=0.038). Comparison of the laboratory and clinical variables between the survivors (n=6) and the deceased (n=13) showed that the survivors had lower SLAM indices at presentation (P=0.004) and APACHE (Acute Physiology, Age, Chronic Health Evaluation) III scores within 24 h after diagnosis of ARDS (P=0.024) than the deceased. The APACHE III scores correlated well with the SLAM indices (r=0.615, P=0.007). Non-SLE patients with ARDS during the study period were selected for comparison to SLE patients with ARDS. Age at the onset of ARDS was younger in SLE (n=19) compared to non-SLE (n=190) (P<0.001). Duration from ARDS onset to death was shorter in SLE patients (P<0. 001). The mortality from ARDS tended to be higher in SLE patients (P=NS). The first-day APACHE III score was significantly higher in deceased SLE patients (n=13) compared to deceased non-SLE patients (n=105) (P=0.001). ARDS was a common premortem event of SLE and showed a high fatality rate in SLE. The most common cause of ARDS in Korean patients with SLE was sepsis by gram negative bacilli. ARDS in SLE developed at a younger age, and progressed more rapidly compared to ARDS in general. The SLAM index and APACHE III score could be useful to predict the prognosis of ARDS in SLE.

Comparison of procalcitonin (PCT) and C-reactive protein (CRP) plasma concentrations at different SOFA scores during the course of sepsis and MODS.

OBJECTIVES: The relation of procalcitonin (PCT) plasma concentrations compared with C-reactive protein (CRP) was analyzed in patients with different severity of multiple organ dysfunction syndrome (MODS) and systemic inflammation. PATIENTS AND METHODS: PCT, CRP, the sepsis-related organ failure assessment (SOFA) score, the Acute Physiology, Age, Chronic Health Evaluation (APACHE) II score and survival were evaluated in 40 patients with systemic inflammation and consecutive MODS over a period of 15 days. RESULTS: Higher SOFA score levels were associated with significantly higher PCT plasma concentrations (SOFA 7-12: PCT 2.62 ng/ml, SOFA 19-24: PCT 15.22 ng/ml) (median), whereas CRP was elevated irrespective of the scores observed (SOFT 7-12: CRP 131 mg/l, SOFT 19-24: CRP 135 mg/l). PCT of non-surviving patients was initially not different from that of survivors but significantly increased after the fourth day following onset of the disease, whereas CRP was not different between both groups throughout the whole observation period. CONCLUSIONS: Measurement of PCT concentrations during multiple organ dysfunction syndrome provides more information about the severity and the course of the disease than that of CRP. Regarding the strong association of PCT and the respective score systems in future studies we recommend evaluation also of the severity of inflammation and MODS when PCT concentrations were compared between different types of disease.

The APACHE III Prognostic System: Risk Prediction of Hospital Mortality for Critically III Hospitalized Adults

The objective of this study was to refine the APACHE (Acute Physiology, Age, Chronic Health Evaluation) methodology in order to more accurately predict hospital mortality risk for critically ill hospitalized adults. We prospectively collected data on 17,440 unselected adult medical/surgical intensive care unit (ICU) admissions at 40 US hospitals (14 volunteer tertiary-care institutions and 26 hospitals randomly chosen to represent intensive care services nationwide). We analyzed the relationship between the patient's likelihood of surviving to hospital discharge and the following predictive variables: major medical and surgical disease categories, acute physiologic abnormalities, age, preexisting functional limitations, major comorbidities, and treatment location immediately prior to ICU admission. The APACHE III prognostic system consists of two options: (1) an APACHE III score, which can provide initial risk stratification for severely ill hospitalized patients within independently defined patient groups; and (2) an APACHE III predictive equation, which uses APACHE III score and reference data on major disease categories and treatment location immediately prior to ICU admission to provide risk estimates for hospital mortality for individual ICU patients. A five-point increase in APACHE III score (range, 0 to 299) is independently associated with a statistically significant increase in the relative risk of hospital death (odds ratio, 1.10 to 1.78) within each of 78 major medical and surgical disease categories. The overall predictive accuracy of the first-day APACHE III equation was such that, within 24 h of ICU admission, 95 percent of ICU admissions could be given a risk estimate for hospital death that was within 3 percent of that actually observed (r2 = 0.41; receiver operating characteristic = 0.90). Recording changes in the APACHE III score on each subsequent day of ICU therapy provided daily updates in these risk estimates. When applied across the individual ICUs, the first-day APACHE III equation accounted for the majority of variation in observed death rates (r2 = 0.90, p<0.0001). The objective of this study was to refine the APACHE (Acute Physiology, Age, Chronic Health Evaluation) methodology in order to more accurately predict hospital mortality risk for critically ill hospitalized adults. We prospectively collected data on 17,440 unselected adult medical/surgical intensive care unit (ICU) admissions at 40 US hospitals (14 volunteer tertiary-care institutions and 26 hospitals randomly chosen to represent intensive care services nationwide). We analyzed the relationship between the patient's likelihood of surviving to hospital discharge and the following predictive variables: major medical and surgical disease categories, acute physiologic abnormalities, age, preexisting functional limitations, major comorbidities, and treatment location immediately prior to ICU admission. The APACHE III prognostic system consists of two options: (1) an APACHE III score, which can provide initial risk stratification for severely ill hospitalized patients within independently defined patient groups; and (2) an APACHE III predictive equation, which uses APACHE III score and reference data on major disease categories and treatment location immediately prior to ICU admission to provide risk estimates for hospital mortality for individual ICU patients. A five-point increase in APACHE III score (range, 0 to 299) is independently associated with a statistically significant increase in the relative risk of hospital death (odds ratio, 1.10 to 1.78) within each of 78 major medical and surgical disease categories. The overall predictive accuracy of the first-day APACHE III equation was such that, within 24 h of ICU admission, 95 percent of ICU admissions could be given a risk estimate for hospital death that was within 3 percent of that actually observed (r2 = 0.41; receiver operating characteristic = 0.90). Recording changes in the APACHE III score on each subsequent day of ICU therapy provided daily updates in these risk estimates. When applied across the individual ICUs, the first-day APACHE III equation accounted for the majority of variation in observed death rates (r2 = 0.90, p<0.0001).